Description of problem: Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info:
Please actually provide a description of the problem when filing bugs. Your summary is insufficient to do anything useful with this bug.
Description of problem: During starting VM: (gnome-boxes:2287): Libvirt.GLib-CRITICAL **: cannot finish stream: stream had I/O failure Version-Release number of selected component (if applicable): libvirt-glib-0.1.6-1.fc19.x86_64 How reproducible: always Steps to Reproduce: 1. Gnome-Boxes -> run VM 2. 3. Actual results: VM is not running. gdb: (gnome-boxes:2287): Libvirt.GLib-CRITICAL **: cannot finish stream: stream had I/O failure /var/log/messages: Apr 22 14:12:10 dhcp131-49 libvirtd[2300]: internal error End of file from monitor Apr 22 14:12:10 dhcp131-49 libvirtd[2300]: port 0 must be in range (5900, 65535) Apr 22 14:12:15 dhcp131-49 libvirtd[2300]: stream had I/O failure Apr 22 14:12:45 dhcp131-49 libvirtd[2300]: stream had I/O failure Expected results: No critical/failure messages. VM is running Additional info:
Other information (before run VM): 1. install new box (e.g. Fedora 18 from live image)(gnome-boxes:3009) GLib-GIO-CRITICAL **: g_file_get_path: assertion `G_IS_FILE (file)' failed (gnome-boxes:3009): Boxes-CRITICAL **: boxes_unattended_installer_remove_disk_from_domain_config: assertion `disk_path != NULL' failed version: gnome-boxes-3.8.1.2-1.fc19.x86_64
running 'G_MESSAGES_DEBUG=Boxes gnome-boxes' from a terminal and attaching the whole log here may give more info Also, make sure "gnome-boxes --checks" is ok.
Created attachment 738936 [details] G_MESSAGES_DEBUG=Boxes gnome-boxes
gnome-boxes --checks • The CPU is capable of virtualization: yes • The KVM module is loaded: yes • Libvirt KVM guest available: yes • Boxes storage pool available: yes • The SELinux context is default: yes Report bugs to <https://bugzilla.gnome.org/enter_bug.cgi?product=gnome-boxes>. Boxes home page: <http://live.gnome.org/Boxes>.
Did you delete 3 VMs when you captured this log, and then tried to start a fedora 18 box?
Yes and installed new one fedora 18 - (but I think deleting has no influence). I have generally problem with starting box.
Does it happen only with f18 or with any OS you try? Is it boxes 3.6 or 3.8?
It is in version of boxes: gnome-boxes-3.8.1.2-1.fc19.x86_64 it happen in F17 too, other OS I did not try.
Today I tried other steps: 1. Downgraded to 3.8.0-1.fc19 version, stopped all VMs (I was not able to do it in 3.8.1.2-1). 2. Update to gnome-boxes-3.8.1.2-1.fc19.x86_64 and run F17, power off, run Fedora 18, power off. And run F17 again, but there was problem: (gnome-boxes:5365): Boxes-DEBUG: libvirt-machine.vala:96: new libvirt machine: fedora17 (gnome-boxes:5365): Boxes-DEBUG: machine.vala:60: State of 'Fedora 17' changed to BOXES_MACHINE_MACHINE_STATE_STOPPED (gnome-boxes:5365): Boxes-DEBUG: libvirt-machine.vala:96: new libvirt machine: fedora18-2 (gnome-boxes:5365): Boxes-DEBUG: machine.vala:60: State of 'Fedora 18 2' changed to BOXES_MACHINE_MACHINE_STATE_STOPPED (gnome-boxes:5365): Boxes-DEBUG: libvirt-machine.vala:96: new libvirt machine: fedora18-3 (gnome-boxes:5365): Boxes-DEBUG: machine.vala:60: State of 'Fedora 18 3' changed to BOXES_MACHINE_MACHINE_STATE_STOPPED (gnome-boxes:5365): Boxes-DEBUG: libvirt-machine.vala:96: new libvirt machine: fedora17-2 (gnome-boxes:5365): Boxes-DEBUG: machine.vala:60: State of 'Fedora 17 2' changed to BOXES_MACHINE_MACHINE_STATE_STOPPED (gnome-boxes:5365): Boxes-DEBUG: libvirt-machine.vala:96: new libvirt machine: fedora18 (gnome-boxes:5365): Boxes-DEBUG: machine.vala:60: State of 'Fedora 18' changed to BOXES_MACHINE_MACHINE_STATE_STOPPED (gnome-boxes:5365): Boxes-DEBUG: machine.vala:60: State of 'Fedora 17' changed to BOXES_MACHINE_MACHINE_STATE_RUNNING (gnome-boxes:5365): Boxes-DEBUG: libvirt-machine.vala:273: enable statistics for Fedora 17 Boxes-Message: machine.vala:147: display Fedora 17 disconnected (gnome-boxes:5365): Boxes-DEBUG: machine.vala:60: State of 'Fedora 17' changed to BOXES_MACHINE_MACHINE_STATE_STOPPED (gnome-boxes:5365): Boxes-DEBUG: libvirt-machine.vala:292: disable statistics for Fedora 17 (gnome-boxes:5365): Boxes-DEBUG: machine.vala:60: State of 'Fedora 18 2' changed to BOXES_MACHINE_MACHINE_STATE_RUNNING (gnome-boxes:5365): Boxes-DEBUG: libvirt-machine.vala:273: enable statistics for Fedora 18 2 (gnome-boxes:5365): GdkPixbuf-WARNING **: GdkPixbufLoader finalized without calling gdk_pixbuf_loader_close() - this is not allowed. You must explicitly end the data stream to the loader before dropping the last reference. (gnome-boxes:5365): Libvirt.GLib-CRITICAL **: cannot finish stream: stream had I/O failure (gnome-boxes:5365): Boxes-DEBUG: machine.vala:60: State of 'Fedora 18 2' changed to BOXES_MACHINE_MACHINE_STATE_STOPPED (gnome-boxes:5365): Boxes-DEBUG: libvirt-machine.vala:292: disable statistics for Fedora 18 2 (gnome-boxes:5365): Boxes-DEBUG: libvirt-machine.vala:523: starting of Fedora 17 was cancelled
(In reply to comment #3) > Other information (before run VM): > 1. install new box (Fedora-18-x86_64-Live-Desktop.iso)(gnome-boxes:3009) > > GLib-GIO-CRITICAL **: g_file_get_path: assertion `G_IS_FILE (file)' failed > > (gnome-boxes:3009): Boxes-CRITICAL **: > boxes_unattended_installer_remove_disk_from_domain_config: assertion > `disk_path != NULL' failed > > version: > gnome-boxes-3.8.1.2-1.fc19.x86_64 1. Start Boxes 2. Click on New 3. Select Fedora-18-x86_64-Live-Desktop.iso from the wizard 4. Start the installation keeping the defaults wizard settings wait until the first reboot during install -> warning in the console. Boxes-Message: machine.vala:147: display Fedora 18 3 disconnected [New Thread 0x7fffe1ada700 (LWP 23335)] [New Thread 0x7fffb3ffc700 (LWP 23336)] [New Thread 0x7fffb47fd700 (LWP 23337)] (gnome-boxes:22014): GLib-GIO-CRITICAL **: g_file_get_path: assertion `G_IS_FILE (file)' failed (gnome-boxes:22014): Boxes-CRITICAL **: boxes_unattended_installer_remove_disk_from_domain_config: assertion `disk_path != NULL' failed
Created attachment 740405 [details] abrt backtrace - Fedora 18 installation and power off. Adding backtrace file generated by abrt when installation of Fedora 18 was finished and Fedora was rebooting/powering off.
All I can tell so far is something really weird going on on your machine it seems. Could you please provide debug log (enabled though setting G_MESSAGES_DEBUG environment variable to 'Boxes').
(In reply to comment #12) > (gnome-boxes:22014): GLib-GIO-CRITICAL **: g_file_get_path: assertion > `G_IS_FILE (file)' failed > > (gnome-boxes:22014): Boxes-CRITICAL **: > boxes_unattended_installer_remove_disk_from_domain_config: assertion > `disk_path != NULL' failed Getting these 2 assertions as well when I restart after doing an install of a live fedora.
(In reply to comment #2) > Description of problem: > During starting VM: (gnome-boxes:2287): Libvirt.GLib-CRITICAL **: cannot > finish stream: stream had I/O failure > > > > Version-Release number of selected component (if applicable): > libvirt-glib-0.1.6-1.fc19.x86_64 > > > How reproducible: > always > > > Steps to Reproduce: > 1. Gnome-Boxes -> run VM > 2. > 3. > > Actual results: > VM is not running. > > gdb: > (gnome-boxes:2287): Libvirt.GLib-CRITICAL **: cannot finish stream: stream > had I/O failure > > /var/log/messages: > Apr 22 14:12:10 dhcp131-49 libvirtd[2300]: internal error End of file from > monitor > Apr 22 14:12:10 dhcp131-49 libvirtd[2300]: port 0 must be in range (5900, > 65535) > Apr 22 14:12:15 dhcp131-49 libvirtd[2300]: stream had I/O failure > Apr 22 14:12:45 dhcp131-49 libvirtd[2300]: stream had I/O failure > > > Expected results: > No critical/failure messages. VM is running > > > Additional info: No deep useful information, but it sounds like duplicate with https://bugzilla.redhat.com/show_bug.cgi?id=950416, the bug was introduced by b2878ed860c, I'm preparing a patch for it.
(In reply to comment #16) > No deep useful information, but it sounds like duplicate with > https://bugzilla.redhat.com/show_bug.cgi?id=950416, the bug was introduced by > b2878ed860c, I'm preparing a patch for it. Oh thanks a lot for the hint, I can hit this bug with both f19 libvirt and jhbuild libvirt (both are 1.0.4). After applying the patch from bug #950416 to the jhbuild build, I no longer get that issue. *** This bug has been marked as a duplicate of bug 950416 ***